READ-ONLY — diagnostic surface for the hardened MarginFi client load. Returns the list of banks the bundled SDK (v6.4.1, IDL 0.1.7) had to skip while fetching the production group, with each record carrying the bank address, best-effort mint + symbol (recovered from raw bytes even when Borsh deco...
AI agents call get_marginfi_diagnostics to retrieve information from VaultPilot MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Even though get_marginfi_diagnostics only reads data, uncontrolled read access leaks sensitive information and racks up API costs — an agent caught in a retry loop can make thousands of calls a minute without anyone noticing.
Risk signalsAdmin/system-level operation
Attacks that exploit this kind of access
READ-ONLY — diagnostic surface for the hardened MarginFi client load. Returns the list of banks the bundled SDK (v6.4.1, IDL 0.1.7) had to skip while fetching the production group, with each record carrying the bank address, best-effort mint + symbol (recovered from raw bytes even when Borsh decode fails), the step where the skip happened (decode / hydrate / tokenData / priceInfo), and the raw error reason. Call this when prepare_marginfi_* reports that a mint you know is listed on mainnet (e.g. USDC) was missed — it will either name the bank explicitly as skipped with the root cause, or confirm the mint truly isn't in the current group. The snapshot reflects the most recent fetchGroupData pass in this process; an empty cache is warmed on demand. No input args. It is categorised as a Read tool in the VaultPilot MCP MCP Server, which means it retrieves data without modifying state.
Register the VaultPilot MCP server in PolicyLayer and add a rule for get_marginfi_diagnostics: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches VaultPilot MCP. Nothing to install.
get_marginfi_diagnostics is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the get_marginfi_diagnostics rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for get_marginfi_diagnostics. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
get_marginfi_diagnostics is provided by the VaultPilot MCP server (vaultpilot-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.